SparseML v0.6.0

New Features:

• YOLOv5 sparsification tutorials and recipes added.
• YOLOv3 sparse transfer learning tutorial added.
• PyTorch image classification using recipes and recipes for ResNet-50 and MobileNet tutorial added.
• BERT additional recipes added for FP32, 3 and 6 layer sparse models.
• Support for phased pruning added: https://arxiv.org/pdf/2106.12379.pdf
• Research folder created for sparsifying passage retrieval.

Changes:

• README updated for Hugging Face transformers integration based on the new implementation.
• ONNX export in PyTorch now supports dictionary inputs.
• Quantized graph export optimizations for YOLOv5.
• PyTorch image classification integration updated to use new manager.modify(...) apis and saves recipes to runs folder.
• DeepSparse YOLO links updated to point at new example location.
• kwargs support added for ONNX export in PyTorch to enable dyanmic_axes and named inputs.

Resolved Issues:

• torch 1.8 quantization export no longer folds incorrectly.
• ONNX toposort issue addressed for nodes with more than two outputs.
• Unused initializers removed in quantized ONNX graphs.

Known Issues:

• None